Abstract
Traits with a complex unknown genetic architecture are common in breeding programs. However, they pose a challenge for selection due to a combination of complex environmental and pleiotropic effects that impede the ability to create mapping populations to characterize the trait's genetic basis. One such trait, seedling emergence of wheat (Triticum aestivum L.) from deep planting, presents a unique opportunity to explore the best method to use and implement genetic selection (GS) models to predict a complex trait. Seventeen GS models were compared using two training populations, consisting of 473 genotypes from a diverse association mapping panel phenotyped from 2015 to 2019 and the other training population consisting of 643 breeding lines phenotyped in 2015 and 2020 in Lind, WA, with 40,368 markers. There were only a few significant differences between GS models, with support vector machines reaching the highest accuracy of 0.56 in a single breeding line trial using cross-validations. However, the consistent moderate accuracy of the parametric models indicates little advantage of using nonparametric models within individual years, but the nonparametric models show a slight increase in accuracy when combing years for complex traits. There was an increase in accuracy using cross-validations from 0.40 to 0.41 using diversity panels lines to breeding lines. Overall, our study showed that breeders can accurately predict and implement GS for a complex trait by using nonparametric machine learning models within their own breeding programs with increased accuracy as they combine training populations over the years.
Abstract from the paper’s OpenAlex record; open copy licensed cc-by-nc-nd. Read at the source ↗
The text the model read, with the source record’s inline markup (italics, sub- and superscripts) rendered and its character entities decoded; quote offsets refer to the raw text.
Claims 6
Claim · model-written summary
Only a few significant differences occurred among GS models.
- Taxon
- wheat
Triticum aestivumt0 - Outcome
- differences between GS modelse0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“There were only a few significant differences between GS models”
t0“wheat (Triticum aestivum L.)”
Claim · model-written summary
Support vector machines attained the highest cross-validation accuracy, 0.56, in one breeding-line trial.
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“support vector machines reaching the highest accuracy of 0.56 in a single breeding line trial using cross-validations”
t0“wheat (Triticum aestivum L.)”
Claim · model-written summary
Parametric models had consistent moderate accuracy.
- Taxon
- wheatTriticum aestivumt0
- Outcome
- moderate accuracye0
- Qualifier
- within individual yearse0
- Comparator
- nonparametric modelse0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“the consistent moderate accuracy of the parametric models indicates little advantage of using nonparametric models within individual years”
t0“wheat (Triticum aestivum L.)”
Claim · model-written summary
Nonparametric models had a slight accuracy increase when years were combined.
- Taxon
- wheatTriticum aestivumt0
- Outcome
- accuracye0
- Direction
- increasee0
- Qualifier
- when combing yearse0
- Comparator
- nonparametric modelse0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“the nonparametric models show a slight increase in accuracy when combing years for complex traits”
t0“wheat (Triticum aestivum L.)”
Claim · model-written summary
Cross-validation accuracy rose from 0.40 to 0.41 using diversity-panel lines to predict breeding lines.
- Taxon
- wheatTriticum aestivumt0
- Outcome
- accuracye0
- Direction
- increasee0
- Value
- 0.40 to 0.41e0
- Comparator
- diversity panels linese0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“There was an increase in accuracy using cross-validations from 0.40 to 0.41 using diversity panels lines to breeding lines.”
t0“wheat (Triticum aestivum L.)”
Claim · model-written summary
The authors concluded that breeders can implement GS for a complex trait with nonparametric models and gain accuracy by combining training populations over years.
- Taxon
- wheatTriticum aestivumt0
- Outcome
- implement GS for a complex traite0
- Qualifier
- as they combine training populations over the yearse0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“breeders can accurately predict and implement GS for a complex trait by using nonparametric machine learning models within their own breeding programs with increased accuracy as they combine training populations over the years.”
t0“wheat (Triticum aestivum L.)”